[Mike Rivers]

I'm looking for something to plot frequency response of EQ settings.
What I'm after is to demonstrate interaction between equalizer bands.
I can put in pink noise and look at (and capture) a spectrum analysis,
but that doesn't look like the curves that you see on spec sheets.

In the past, when I've wanted this, I've used a generator, swept it
through the frequency range, recorded selected points in Excel, and
then used Excel's graphing tools to make a plot. I've just never been
able to fully manage Excel, though, and can never seem to get the
range right (and consistent with every plot - it has a mind of its own
that I haven't learned to control).

What I'm dreaming of is something that's the equivalent of an old
motor-driven General Radio audio oscillator connected (through the
equipment) to a graphic plotter. Seems like that should be easy enough
to do in software but I've just never found anything to do it other
than the Prism or Audio Precision analyzers, which involve expensive
hardware.

Anyone have an idea? Taking and recording the data by hand isn't a big
deal, I just have a problem making nice looking (publication-quality)
plots of dB vs frequency on virtual semilog graph paper. And, yes,
I've done plots on graph paper and scanned them, but aren't we in the
computer age yet?

[Preben Friis]

"Mike Rivers" wrote
I'm looking for something to plot frequency response of EQ settings.
What I'm after is to demonstrate interaction between equalizer bands.
I can put in pink noise and look at (and capture) a spectrum analysis,
but that doesn't look like the curves that you see on spec sheets.

Have you looked at RMAA?
http://audio.rightmark.org/index_new.shtml

It is able to plot curves like this one:
http://fierytrance.ru/img/UB502/Line1EQfr.png

/Preben Friis

[Mike Rivers]

On Dec 22, 8:21 am, "Preben Friis" wrote:

Have you looked at RMAA?http://audio.rightmark.org/index_new.shtml

Yes. I have it loaded on a couple of computers but I have continual
problems getting it to work right. It seems like it'll work with the
built-in sound card on my laptop (and that's probably OK for the
purpose) but it won't work with any of my better quality Firewire
interfaces or my Digigram VX Pocket card.

I haven't tried it for several months. Maybe there's a newer version
that works better. Seems like I recall that the paid-for version
claims to support ASIO drivers, but I don't want to buy it until I
KNOW it'll work for me.

[Scott Dorsey]

Mike Rivers wrote:
What I'm dreaming of is something that's the equivalent of an old
motor-driven General Radio audio oscillator connected (through the
equipment) to a graphic plotter. Seems like that should be easy enough
to do in software but I've just never found anything to do it other
than the Prism or Audio Precision analyzers, which involve expensive
hardware.

You are six months too late... I just donated my B&K 2112 motor-driven
analyzer to a museum this summer. I'd have happily given it to you too.

I replaced it with a Rockland 5830 FFT analyzer. It also does swept
sine measurements. You're welcome to borrow it any time.

Anyone have an idea? Taking and recording the data by hand isn't a big
deal, I just have a problem making nice looking (publication-quality)
plots of dB vs frequency on virtual semilog graph paper. And, yes,
I've done plots on graph paper and scanned them, but aren't we in the
computer age yet?

To be honest, I mostly play a swept sine test CD and look at how much
the meters bounce, unless I really need qualitative measurements for
a customer or something.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Mike Rivers]

On Dec 22, 10:41 am, (Scott Dorsey) wrote:

You are six months too late... I just donated my B&K 2112 motor-driven
analyzer to a museum this summer.

Oh, no! Now your boat will drift away. g

If there's anything interesting to be seen, I'd like to be able to
show it in a review rather than just describe it in words.

[Preben Friis]

"Mike Rivers" wrote
On Dec 22, 8:21 am, "Preben Friis" wrote:

Have you looked at RMAA?http://audio.rightmark.org/index_new.shtml

Yes. I have it loaded on a couple of computers but I have continual
problems getting it to work right. It seems like it'll work with the
built-in sound card on my laptop (and that's probably OK for the
purpose) but it won't work with any of my better quality Firewire
interfaces or my Digigram VX Pocket card.

It is possible to use RMAA in "offline" mode. You generate a test file, and
then use your favorite program to loop it through the device under test.
Then all what is left is to load the resulting file back into RMAA. That is
the way I have used when testing Digidesign hardware, which does not work
directly with the free RMAA either. It does require some extra time, but it
beats entering numbers in Excel anytime.

/Preben Friis

[Ron Capik]

Mike Rivers wrote:

I'm looking for something to plot frequency response of EQ settings.
What I'm after is to demonstrate interaction between equalizer bands.
I can put in pink noise and look at (and capture) a spectrum analysis,
but that doesn't look like the curves that you see on spec sheets.
...snip.....

Hmmmm, I wrote a program that does what you want to do, and even
used it to look at EQ phase and frequency response. I tend to use
it with wav files because the sound card interface tends to be a bit
flaky. It does a comparison [a complex difference in the frequency
domain] of one channel with the other.
It can also find echoes, time offsets, plot channel correlation and
has an XY scope of sorts. [I've never been able to get a program to
to truly mimic a real XY scope. Can't seem to emulate the persistance
effects of a scope tube.]

Thing is, it's in a rather uncommon language [LabVIEW] and it's not
really ready for prime time. It's more of a tech toy for my enjoyment. G

However, I do believe SMAART can do similar things.


Later...

Ron Capik
--

[Mike Rivers]

On Dec 22, 11:58 am, "Preben Friis" wrote:

It is possible to use RMAA in "offline" mode. You generate a test file, and
then use your favorite program to loop it through the device under test.

That works, but it's as much trouble as doing it manually. At least
that's the way it is for me.

[Peter Larsen[_2_]]

Mike Rivers wrote:

I'm looking for something to plot frequency response of EQ settings.
What I'm after is to demonstrate interaction between equalizer bands.
I can put in pink noise and look at (and capture) a spectrum analysis,
but that doesn't look like the curves that you see on spec sheets.

There is a loudspeaker design application out there that may be just what
you are looking for, it comes with a measurement engine. Take a look at:

http://www.audua.com/


Kind regards

Peter Larsen

[Mike Rivers]

On Dec 22, 8:21 am, "Preben Friis" wrote:

Have you looked at RMAA?

I don't know what it is about this program, but I just can't get it to
work with the audio hardware I have around here. I just tried again
this afternoon, version 6.5 and 6.something, using a Mackie Satellite
and a Digigram VX Pocket card and just kept getting various errors,
some of which completely locked up the program. It seems to work with
the built-in sound card (SigmaTel) but that has only a mic level
input, so I have to run tests at mic level.

When the Satellite decides to sort of work, I can't get the level set
correctly. It wants to be about -1 dBFS, but no matter how I adjust
input and output levels, either it runs about -3 dBFS (and complains).
Anything above that clips the Satellite input.

It does exactly what I'm looking for, if only it did it long enough to
be useful.

Grrrrrr!

[Arny Krueger]

"Mike Rivers" wrote in message


On Dec 22, 11:58 am, "Preben Friis"
wrote:

It is possible to use RMAA in "offline" mode. You
generate a test file, and then use your favorite program
to loop it through the device under test.

I do that quite frequently.

That works, but it's as much trouble as doing it
manually.

Surely, you jest.

At least that's the way it is for me.

You must have rocket-powered fingers for plotting curves from dozens of
points.

[Don P.]

Mike Rivers scribbled:

On Dec 22, 8:21 am, "Preben Friis" wrote:

Have you looked at RMAA?

I don't know what it is about this program, but I just can't get it to
work with the audio hardware I have around here. I just tried again
this afternoon, version 6.5 and 6.something, using a Mackie Satellite
and a Digigram VX Pocket card and just kept getting various errors,
some of which completely locked up the program. It seems to work with
the built-in sound card (SigmaTel) but that has only a mic level
input, so I have to run tests at mic level.

When the Satellite decides to sort of work, I can't get the level set
correctly. It wants to be about -1 dBFS, but no matter how I adjust
input and output levels, either it runs about -3 dBFS (and complains).
Anything above that clips the Satellite input.

It does exactly what I'm looking for, if only it did it long enough to
be useful.

Grrrrrr!


You know, another thing I've done is to just use Audacity to generate say
4 or 5 minutes of white noise, save that as a wave file, burn it to an
audio CD, then run that through whatever system, and recapture the
resulting audio. Import that back into Audacity and use it's spectrum
display to show the results. Also comes in handy for audio files to see
if they're band-limited, and for checking noise floors of recording
devices.

I used this method once to check a cassette recorder. Played the CD onto
the cassette, then recorded the output of the tape back to a CD-RW (so I
didn't "use up" a blank), then rip that back into the computer as a wave
file, and analyze in Audacity. Cheezy,but it works!

[Frank Vuotto]

I came in late so I don't know if it has been mentioned that you can
get a 30 day version of Smaart v.6 he

http://www.eaw.com/products/software.../download.html

This version works with most soundcards(at last) but, in my opinion,
they really screwed the interface.

Frank /~ http://newmex.com/f10
@/
---- Posted via Pronews.com - Premium Corporate Usenet News Provider ----
http://www.pronews.com offers corporate packages that have access to 100,000+ newsgroups

[Mike Rivers]

On Dec 23, 8:53 pm, "Arny Krueger" wrote:

You must have rocket-powered fingers for plotting curves from dozens of
points.

I didn't say doing it manually would be faster, but it's less
frustrating, at least for me. I don't understand why such a well
regarded and often recommended program simply doesn't work for me, but
then I'm no software expert.

[Mike Rivers]

On Dec 23, 9:12 pm, "Don P." wrote:
You know, another thing I've done is to just use Audacity to generate say
4 or 5 minutes of white noise, save that as a wave file, burn it to an
audio CD, then run that through whatever system, and recapture the
resulting audio. Import that back into Audacity and use it's spectrum
display to show the results.

A spectrum disply of noise filtered by the system under test isn't
stable enough to be able to examine small differences in frequency
response. I have spectrum analyzers that work as spectrum analyzers,
but they aren't what I want here.

I haven't tried the record-and-play procedure with RightMark, mostly
because I don't understand why that would work and using it as a
"thru" system won't. It's still talking (or rather, not talking) to
the same hardware. Maybe it will work. Maybe I will try it. But I
like things that wok the way they're supposed to work and don't trust
those that need work-arounds.

[Laurence Payne]

On Sun, 23 Dec 2007 20:19:58 -0800 (PST), Mike Rivers
wrote:

I don't understand why such a well
regarded and often recommended program simply doesn't work for me, but
then I'm no software expert.

Didn't you tell us you were using a demo version that doesn't support
ASIO drivers?

[Mike Rivers]

On Dec 24, 6:27 am, Laurence Payne NOSPAMlpayne1ATdsl.pipex.com
wrote:

Didn't you tell us you were using a demo version that doesn't support
ASIO drivers?

Yes, but I'm not trying to use it with ASIO drivers. It's supposed to
work with WDM drivers and that's what I'm using - at least that's what
I'm choosing in the program. I assume that's what it's actually using.
I can tell Sound Forge or Audacity to use WDM drivers for the same
hardware and they work fine. RightMark's test mode tells me that the
hardware is working and is capable of full duplex operation. It just
doesn't work when running the test.

Given the number of suggestions that I use it in the two-stage mode
(recording the test signal and saving a WAV file, then playing the
recording for analysis) I suspect that I'm not the only one having a
problem with RMMA working the way it should.

[Arny Krueger]

"Mike Rivers" wrote in message

On Dec 24, 6:27 am, Laurence Payne
NOSPAMlpayne1ATdsl.pipex.com wrote:

Didn't you tell us you were using a demo version that
doesn't support ASIO drivers?

Yes, but I'm not trying to use it with ASIO drivers. It's
supposed to work with WDM drivers and that's what I'm
using - at least that's what I'm choosing in the program.
I assume that's what it's actually using. I can tell
Sound Forge or Audacity to use WDM drivers for the same
hardware and they work fine. RightMark's test mode tells
me that the hardware is working and is capable of full
duplex operation. It just doesn't work when running the
test.

Have you ever seen it work?

Between PCAVTech and life, I've tested more than 100 audio interfaces.
Setting up a loop-back test is one of those things that is conceptually
simple, but sometimes takes more hassle than imaginable. If you're lucky,
your first try is one of the easy ones, so then when you get a hard one, you
know that *it has to work, I'm just not doing something right*.

Given the number of suggestions that I use it in the
two-stage mode (recording the test signal and saving a
WAV file, then playing the recording for analysis) I
suspect that I'm not the only one having a problem with
RMMA working the way it should.

The two-stage mode is for testing things you can't loop back, like say a CD
player or the record/play of a 2-headed cassette recorder.

BTW, that's a recommended procedure for really appreciating digital - test a
CD player and then test a cassette recorder and see how far consumer audio
has come! It brings to mind the sad story of a well-known high end audio
guru who tried to bring the cassette format back a couple of years back.

But I admit, I've tested more than a few audio interfaces that way, using
Cool Edit 2.1 as my record/play manager.

[Mike Rivers]

On Dec 24, 9:11 am, "Arny Krueger" wrote:

Have you ever seen it work?

Yes, with the built-in sound card on one of my laptop computers, but
I'd prefer not to use that for measurements.

Setting up a loop-back test is one of those things that is conceptually
simple, but sometimes takes more hassle than imaginable. If you're lucky,
your first try is one of the easy ones, so then when you get a hard one, you
know that *it has to work, I'm just not doing something right*.

So what am I not doing right?

The two-stage mode is for testing things you can't loop back, like say a CD
player or the record/play of a 2-headed cassette recorder.

But I AM testing something I can loop back. I can't even get it to
work when I'm "testing" a cable. It was suggested, maybe even at you
at one time, that when a loopback test doesn't work, a two-step test
might work.

[Arny Krueger]

"Mike Rivers" wrote in message

On Dec 24, 9:11 am, "Arny Krueger"
wrote:

Have you ever seen it work?

Yes, with the built-in sound card on one of my laptop
computers, but I'd prefer not to use that for
measurements.

To summarize, you got a loop-back test to work with your laptop.

Setting up a loop-back test is one of those things that
is conceptually simple, but sometimes takes more hassle
than imaginable. If you're lucky, your first try is one
of the easy ones, so then when you get a hard one, you
know that *it has to work, I'm just not doing something
right*.

So what am I not doing right?

So, you're saying you can't get your Digigram to loop back through a cable?

The two-stage mode is for testing things you can't loop
back, like say a CD player or the record/play of a
2-headed cassette recorder.

But I AM testing something I can loop back.

Well, so you may think. You may have been mislead. Not everything that RMAA
says will loop in its preliminary tests will actually do so when the
electrons hit the copper. Sad but true.

I can't even get it to work when I'm "testing" a cable.

We can presume that:

(1) Given the right audio interface, RMAA works, because it woprked with
your Laptop.

(2) The ability of copper wire to loopback is exceedingly high. ;-)

We then conclude that the problem may be:

(1) With the audio interface you are now using.

It was suggested, maybe even at you at one time, that when a
loopback test doesn't work, a two-step test might work.

That's my current point. Just because it is a modern audio interface and
records and plays doesn't mean that it can loop.

Try testing something that doesn't have to loop, like maybe your digital
recorder.

Try holding your nose, and doing some tests with the audio interface in the
laptop.

[Mike Rivers]

On Dec 24, 10:12 am, "Arny Krueger" wrote:

To summarize, you got a loop-back test to work with your laptop.

To update, I got the program to work occasionally with my laptop. I
did some more experimenting with it this morning and while I could get
signals in and out of the mixer I was testing, I never finished the
job.

I thought that the planets were all aligned and I was getting some
results. With the mixer's channel EQ set flat, I got a relatively
straight line for frequency response (I was only testing frequency
response). I never got the levels set so that the happiness box was
green, but got the test to run anyway. I was setting the level so
that the 1 kHz peak on the graph on the level-setting screen was at
-12 dB to allow some headroom when I twiddled with the EQ knobs.

So I saved the flat frequency response run, set the HF EQ for a boost
of 6 dB (I calibrated it first) and it told me that some fairly small
percentage of the samples was clipped, so I backed off the level and
did another run, This time it worked, but the reference level
(compared to EQ set flat) had changed. In another run, I had a mid-
range cut and even though I was cutting (and could see it on the
mixer's output meters) it complained that there was clipping.

So I went back to flat, ran the test setting the level with the 1 kHz
peak at -16 dB. This time when I got to the mid-range cut, it aborted
the test, said that it lost the sync signal and that RMMA had to quit
(which it did, losing my previous test runs) That's a badly behaved
program, for sure.

This is the sort of nonsense up with which I shall not put.

So, you're saying you can't get your Digigram to loop back through a cable?

That's a different issue, but basically, yes. I run the test tone, I
see the meters on the Digigram mixer control panel moving, I see the
meters on Rightmark moving, There's about 2 dB difference between the
signal level being too low, the signal level being too high, and when
it's neither too low or too high, the Inter-channel leakage indicator
is red (actually, it's always red) and the "problem" box says it looks
like I have the channels interchanged. I don't, but when I swap the
cables, the symptom doesn't change. Something isn't talking right.
Sound Forge knows left from right and knows that they're not reversed.

But I AM testing something I can loop back.

Well, so you may think. You may have been mislead. Not everything that RMAA
says will loop in its preliminary tests will actually do so when the
electrons hit the copper. Sad but true.

So what are the exceptions? Anything known or predictable? Or is it
just a matter of what works works and what doesn't, doesn't? That's
kind of a good description of a bad program.

Try testing something that doesn't have to loop, like maybe your digital
recorder.

That's not what I want to test. I suppose it might be educational to
try using one computer as the generator and another computer as the
test instrument, running one in the "record the test" mode and the
other in the "test from playback" mode. But even knowing that this
works doesn't make it useful to me.

Try holding your nose, and doing some tests with the audio interface in the
laptop.

I even stood on one foot with my candle lit and my umbrella open.
Still no go.